LABORATORIO DE URGENCIAS

Ferritin and Diagnosis of Iron Deficiency without Anemia

Levy Jäger, Yael Rachamin, Oliver Senn et al.
          Institute of Primary Care, University Hospital Zurich, University of Zurich, Zurich, Switzerland
            Campus Stiftung Lindenhof Bern (SLB), Bern, Switzerland

 JAMA Network Open. 2024;7(8):e2425692

Summarized by: Dra. Marta Papponetti

Iron deficiency is a common condition and leading cause of years of life with disability worldwide, primarily due to subsequent anemia.¹
Decreased iron stores that do not yet result in anemia, simply referred to as non-anemic iron deficiency, have recently gained importance as a distinct clinical entity.² Associated with a variety of symptoms, including fatigue, restless legs syndrome, and hair loss,³ non-anemic iron deficiency has been estimated to be more common than anemic iron deficiency.²
Measuring (serum) ferritin is considered the mainstay of diagnosing iron deficiency. Ferritin is a frequently ordered laboratory test in various high-resource primary care settings.²-¹ Notably, a Swiss study found that 27% of the population underwent serum ferritin testing in 2018,¹ and ferritin was among the most frequently ordered laboratory tests in Swiss primary care between 2009 and 2018.²
Guidelines on iron deficiency are conflicting regarding which populations benefit most from ferritin testing.² For example:
* The US Preventive Services Task Force specifically mentions children and pregnant women, but does not offer any recommendations for the general population.
* A major Swiss primary care guideline advises against screening for iron deficiency in the general population, with exceptions for conditions that require special consideration due to increased risk, such as inflammatory bowel disease (IBD), or when evidence suggests a benefit of having complete iron stores, such as chronic kidney disease (CKD).11
Optimal ferritin cutoff values ​​for diagnosing iron deficiency, especially non-anemic iron deficiency, are controversial. Different guidelines suggest widely varying cutoff values, ranging from 12 to 15 ng/mL, through 25 to 30 ng/mL, to 45 to 50 ng/mL in the general population (to convert to micrograms per liter, multiply by 1).4 The choice of the ferritin cutoff value can have important implications. On the one hand, choosing too low a cutoff value could result in discontinuing treatment in patients affected by the negative health consequences of iron deficiency. On the other hand, choosing too high a cutoff value could lead to overtreatment of patients who do not benefit from iron therapy, with potential adverse effects (especially with oral preparations12) and an unjustified waste of healthcare resources.
Previous studies have estimated the incidence of anemic iron deficiency in primary care,13,14 but have not evaluated the choice of different ferritin cutoff values, nor has the incidence of non-anemic iron deficiency been investigated. Adequate data would be essential for a comprehensive evaluation of ferritin testing in primary care. Our study aims to address these shortcomings by estimating how the incidence of anemic and non-anemic iron deficiency diagnoses depends on the choice of ferritin cutoff value and by examining the determinants of ferritin testing in Swiss primary care.

Objective
To investigate the association of ferritin threshold selection with the incidence of anemic and non-anemic iron deficiency diagnoses in primary care.
A retrospective cohort study evaluating patients aged ≥18 years with at least one consultation with a general practitioner participating in the Family Medicine Recovery Using Electronic Medical Records (FIRE) project, a Swiss primary care electronic medical record database, from January 1, 2021, to November 30, 2023, was performed.
Sex, age, patient clinical characteristics, and GP professional characteristics were assessed.

Results
The study included 255,351 patients (mean age 52 years; 52.1% women).
Per 1,000 patient-years, with ferritin thresholds of 15, 30, and 45 mg/L, iron deficiency diagnoses had incidences of 10.9; 29.9 and 48.3 cases, respectively.
Diagnoses of non-anemic iron deficiency had incidences of 4.1, 14.6, and 25.8 cases, respectively. Diagnoses of anemic iron deficiency had incidences of 3.5, 6.0, and 7.5 cases, respectively.
Ferritin testing showed notable associations with fatigue, anemia, and iron therapy. Ferritin testing was associated with female sex in all age groups, including postmenopausal women. Seventy-two point one percent of patients who underwent ferritin testing had concomitant hemoglobin determinations, and 49.6% had C-reactive protein determinations.

Discussion
This study of more than 255,000 patients investigated the determinants and variation of ferritin testing and the incidence of iron deficiency diagnoses in Swiss primary care.
A significant association between the choice of ferritin threshold and iron deficiency diagnosis rates was observed, especially for non-anemic iron deficiency. Furthermore, a substantial degree of variation in ferritin testing was found, along with interesting associations, such as higher testing rates in postmenopausal women compared with age-matched men, and higher rates of testing ordered by female and younger general practitioners.
Gaps in the quality of ferritin testing were also found, with a large proportion ordered without concomitant hemoglobin and CRP measurements.
The authors state that their most compelling finding regarding the incidence of iron deficiency diagnoses was its close dependence on the choice of ferritin threshold, especially for non-anemic iron deficiency.
The use of ferritin thresholds as treatment thresholds has been widely debated, and a recent Cochrane review found insufficient evidence to recommend a specific threshold in a healthy, asymptomatic population. Even within Switzerland, conflicting recommendations exist. A guideline from a major Swiss primary care network emphasizes that iron replacement is not justified in the presence of ferritin concentrations >15 mg/L, whereas a local expert panel has recommended diagnosing iron deficiency in the presence of ferritin concentrations <30 mg/L. The authors maintain that the results obtained in this study show that these controversies affect the management of a considerable number of patients.
The incidence of anemic iron deficiency diagnoses of approximately 13 cases/1000 patient-years is comparable to the results of a similar multinational European study. “We are unaware of other studies that have estimated iron deficiency diagnosis rates based on routine data in high-resource settings,” the authors write. “However, the lack of concomitant hemoglobin and CRP measurements has important implications for the interpretation of such incidences.”
The proportion of ferritin tests without concomitant CRP or hemoglobin testing was surprisingly high because most guidelines explicitly recommend screening for systemic inflammation and anemia in iron deficiency workup.
Hemoglobin and CRP tests were associated with fatigue and fewer prior primary care visits, suggesting that they were used more frequently in patients whose iron deficiency was screened based on symptoms rather than episodes of care involving routine screening.
Often, ferritin deficiency may have prompted ferritin testing. Many of the symptoms, especially fatigue, are nonspecific but are more common in women and present concerns in primary care. Although adjustments were made for fatigue, its low frequency of documentation may have resulted in residual confounding that partially explains the persistent association with female sex.
Notable associations of ferritin testing with GP characteristics were also observed, with female GPs and younger GPs being more likely to perform ferritin testing. The gender difference can be interpreted in the context of previous findings that female GPs provide more preventive care than their male colleagues, and the age disparity may be related to variation in the information-seeking behavior of GPs of different ages.
The GP-level variation in ferritin testing expressed by the mean hazard ratio was surprising because it was comparable to several adjusted hazard ratios that expressed associations with clinical factors. This finding is consistent with previously observed evidence of unjustified variation in the use of ferritin tests across primary care settings, which has been interpreted as an indicator of overuse.
Most ferritin tests were not accompanied by requests for additional iron studies, suggesting that GPs were aware of the recommendation to use ferritin as a first-line test for iron deficiency. This result contrasts with findings from other countries, such as Australia and Spain, where other iron studies were overordered compared to ferritin.
On the other hand, we observed that serum iron was the most frequently requested additional iron study, contrary to the recommendation of local guidelines to avoid its use. This finding suggests a knowledge gap regarding the use of iron studies, in line with results from the international literature.
In summary, they say, “our findings can be understood in the context of the clinical uncertainty faced by GPs regarding the diagnosis of iron deficiency.” This uncertainty may be due in part to the lack of consensus among different recommendations for iron deficiency screening and ferritin thresholds, ultimately necessitating greater guidance on iron deficiency treatment in primary care.

Limitations
This study has some limitations. The FIRE database does not allow access to presenting symptoms as documented in medical records, which is a major limitation, along with the lack of referral information. Furthermore, access to information on gastrointestinal risk factors such as celiac disease was limited. However, information on patients' diet, which could have prompted iron deficiency screening, was also unavailable. A previous study found no significant differences in the prevalence of iron deficiency among Swiss omnivores, vegetarians, or vegans.

Conclusion
This study demonstrates a substantial increase in the rate of iron deficiency diagnoses when ferritin thresholds of 30 and 45 mg/L, respectively, are chosen instead of 15 mg/L. The results provide an information base for health systems' evaluation of ferritin testing in primary care. Furthermore, the need to agree on guidelines for the diagnosis of iron deficiency in primary care is highlighted.

Referencias
1.GBD 2021 Anaemia Collaborators.  Prevalence, years lived with disability, and trends in anaemia burden by severity and cause, 1990-2021: findings from the Global Burden of Disease Study 2021.   Lancet Haematol. 2023;10(9):e713-e734. doi:10.1016/S2352-3026(23)00160-6PubMedGoogle ScholarCrossref
2.Al-Naseem  A, Sallam  A, Choudhury  S, Thachil  J.  Iron deficiency without anaemia: a diagnosis that matters.   Clin Med (Lond). 2021;21(2):107-113. doi:10.7861/clinmed.2020-0582PubMedGoogle ScholarCrossref
3.Beatrix  J, Piales  C, Berland  P, Marchiset  E, Gerbaud  L, Ruivard  M.  Non-anemic iron deficiency: correlations between symptoms and iron status parameters.   Eur J Clin Nutr. 2022;76(6):835-840. doi:10.1038/s41430-021-01047-5PubMedGoogle ScholarCrossref
4.Peyrin-Biroulet  L, Williet  N, Cacoub  P.  Guidelines on the diagnosis and treatment of iron deficiency across indications: a systematic review.   Am J Clin Nutr. 2015;102(6):1585-1594. doi:10.3945/ajcn.114.103366PubMedGoogle ScholarCrossref
5. O’Sullivan  JW, Stevens  S, Hobbs  FDR,  et al.  Temporal trends in use of tests in UK primary care, 2000-15: retrospective analysis of 250 million tests.   BMJ. 2018;363:k4666. doi:10.1136/bmj.k4666PubMedGoogle ScholarCrossref
6. Salinas  M, López-Garrigós  M, Flores  E, Leiva-Salinas  C; Pilot Group of the Appropriate Utilization of Laboratory Tests (REDCONLAB) working group.  Primary care requests for anaemia chemistry tests in Spain: potential iron, transferrin and folate over-requesting.   J Clin Pathol. 2017;70(9):760-765. doi:10.1136/jclinpath-2016-204249 
7. Nguyen  LT, Guo  M, Hemmelgarn  B,  et al.  Evaluating practice variance among family physicians to identify targets for laboratory utilization management.   Clin Chim Acta. 2019;497:1-5. doi:10.1016/j.cca.2019.06.017PubMedGoogle ScholarCrossref
8. Huber  CA, Nagler  M, Rosemann  T, Blozik  E, Näpflin  M, Markun  S.  Trends in micronutrient laboratory testing in Switzerland: a 7-year retrospective analysis of healthcare claims data.   Int J Gen Med. 2020;13:1341-1348. doi:10.2147/IJGM.S275406PubMedGoogle ScholarCrossref
9. Schumacher  LD, Jäger  L, Meier  R,  et al.  Trends and between-physician variation in laboratory testing: a retrospective longitudinal study in general practice.   J Clin Med. 2020;9(6):1787. doi:10.3390/jcm9061787PubMedGoogle ScholarCrossref
10. Kemper  AR, Fan  T, Grossman  DC, Phipps  MG.  Gaps in evidence regarding iron deficiency anemia in pregnant women and young children: summary of US Preventive Services Task Force recommendations.   Am J Clin Nutr. 2017;106(suppl 6):1555S-1558S. doi:10.3945/ajcn.117.155788PubMedGoogle ScholarCrossref
11. Chmiel  C, Beise  U. mediX-Guidelines: Eisenmangel. Verein mediX. 2023. Accessed February 9, 202 https://www.medix.ch/wissen/guidelines/eisenmangel/         
12. Tolkien  Z, Stecher  L, Mander  AP, Pereira  DI, Powell  JJ.  Ferrous sulfate supplementation causes significant gastrointestinal side-effects in adults: a systematic review and meta-analysis.   PLoS One. 2015;10(2):e0117383. doi:10.1371/journal.pone.0117383PubMedGoogle ScholarCrossref
13.Levi  M, Rosselli  M, Simonetti  M,  et al.  Epidemiology of iron deficiency anaemia in four European countries: a population-based study in primary care.   Eur J Haematol. 2016;97(6):583-593. doi:10.1111/ejh.12776PubMedGoogle ScholarCrossref
14.+Levi  M, Simonetti  M, Marconi  E,  et al.  Gender differences in determinants of iron-deficiency anemia: a population-based study conducted in four European countries.   Ann Hematol. 2019;98(7):1573-1582. doi:10.1007/s00277-019-03707-wPubMedGoogle ScholarCrossref
15.Chmiel  C, Bhend  H, Senn  O, Zoller  M, Rosemann  T; FIRE study-group.  The FIRE project: a milestone for research in primary care in Switzerland.   Swiss Med Wkly. 2011;140:w13142. doi:10.4414/smw.2011.13142PubMedGoogle ScholarCrossref
16.World Health Organization Collaborating Centre for Drug Statistics Methodology. ATC classification index with DDDs. 2023. Accessed January 4, 2024. https://www.whocc.no/
17.World Organization of National Colleges, Academies and Academic Associations of General Practitioners/Family Physicians (WONCA). ICPC-2-R: International Classification of Primary Care. Oxford University Press; 2005. Accessed July 16, 2024. http://www.ph3c.org/PH3C/docs/27/0000908.pdf
18.Federal Act on Research involving Human Beings 30 September 2011 (Status as of 01 January 2014). 2011. Updated September 30, 2011. Accessed February 9, 2024.  https://www.fedlex.admin.ch/eli/cc/2013/617/de
19.Benchimol  EI, Smeeth  L, Guttmann  A,  et al; RECORD Working Committee.  The Reporting of Studies Conducted Using Observational Routinely-Collected Health Data (RECORD) statement.   PLoS Med. 2015;12(10):e1001885. doi:10.1371/journal.pmed.1001885PubMedGoogle ScholarCrossref
20.World Health Organization. Haemoglobin concentrations for the diagnosis of anaemia and assessment of severity. World Health Organization. 2011. Accessed February 9, 2024. https://apps.who.int/iris/handle/10665/85839
21.Gemeindetypologie und Stadt/Land-Typologie 2012. Swiss Federal Statistical Office. 2017. Accessed January 4, 2024. https://www.bfs.admin.ch/bfs/de/home/statistiken/kataloge-datenbanken/karten.assetdetail.2543323.html
22.R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. 2023. Accessed July 16, 2024. https://www.R-project.org/
23.Nowak  A, Angelillo-Scherrer  A, Betticher  D,  et al.  Swiss Delphi study on iron deficiency.   Swiss Med Wkly. 2019;149:w20097. doi:10.4414/smw.2019.20097PubMedGoogle ScholarCrossref
24.Naveed  K, Goldberg  N, Shore  E,  et al.  Defining ferritin clinical decision limits to improve diagnosis and treatment of iron deficiency: a modified Delphi study.   Int J Lab Hematol. 2023;45(3):377-386. doi:10.1111/ijlh.14016PubMedGoogle ScholarCrossref
25.Fletcher  A, Forbes  A, Svenson  N, Wayne Thomas  D; A British Society for Haematology Good Practice Paper.  Guideline for the laboratory diagnosis of iron deficiency in adults (excluding pregnancy) and children.   Br J Haematol. 2022;196(3):523-529. doi:10.1111/bjh.17900PubMedGoogle ScholarCrossref
26.Snook  J, Bhala  N, Beales  ILP,  et al.  British Society of Gastroenterology guidelines for the management of iron deficiency anaemia in adults.   Gut. 2021;70(11):2030-2051. doi:10.1136/gutjnl-2021-325210PubMedGoogle ScholarCrossref
27.Short  MW, Domagalski  JE.  Iron deficiency anemia: evaluation and management.   Am Fam Physician. 2013;87(2):98-104.PubMedGoogle Scholar
28.Ulm  K.  A simple method to calculate the confidence interval of a standardized mortality ratio (SMR).   Am J Epidemiol. 1990;131(2):373-375. doi:10.1093/oxfordjournals.aje.a115507PubMedGoogle ScholarCrossref
29.Morgen  EK, Naugler  C.  Inappropriate repeats of six common tests in a Canadian city: a population cohort study within a laboratory informatics framework.   Am J Clin Pathol. 2015;144(5):704-712. doi:10.1309/AJCPYXDAUS2F8XJYPubMedGoogle ScholarCrossref
30.Fisher  LD, Lin  DY.  Time-dependent covariates in the Cox proportional-hazards regression model.   Annu Rev Public Health. 1999;20:145-157. doi:10.1146/annurev.publhealth.20.1.145PubMedGoogle ScholarCrossref
31.Coxme: Mixed Effects Cox Models (R package version 2.2-18.1). 2022. Accessed July 16, 2024. https://CRAN.R-project.org/package=coxme
32.van der Leeden R, Meijer E, Busing FMTA. Resampling multilevel models. In: Leeuw  Jd, Meijer  E, eds.  Handbook of Multilevel Analysis. Springer; 2008:401-433. doi:10.1007/978-0-387-73186-5
33.Austin  PC, Wagner  P, Merlo  J.  The median hazard ratio: a useful measure of variance and general contextual effects in multilevel survival analysis.   Stat Med. 2017;36(6):928-938. doi:10.1002/sim.7188PubMedGoogle ScholarCrossref
34.Bates  D, Mächler  M, Bolker  B, Walker  S.  Fitting linear mixed-effects models using lme4.   J Stat Softw. 2015;67:1-48. doi:10.18637/jss.v067.i01Google ScholarCrossref
35.Daru  J, Colman  K, Stanworth  SJ, De La Salle  B, Wood  EM, Pasricha  SR.  Serum ferritin as an indicator of iron status: what do we need to know?   Am J Clin Nutr. 2017;106(suppl 6):1634S-1639S. doi:10.3945/ajcn.117.155960PubMedGoogle ScholarCrossref
36.Garcia-Casal  MN, Pasricha  SR, Martinez  RX, Lopez-Perez  L, Peña-Rosas  JP.  Serum or plasma ferritin concentration as an index of iron deficiency and overload.   Cochrane Database Syst Rev. 2021;5(5):CD011817.PubMedGoogle Scholar
37.NPS MedicineWise. General practice insights report July 2016-June 2017: a working paper. 2018. Accessed May 24, 2024. https://www.safetyandquality.gov.au/our-work/indicators-measurement-and-reporting/medicineinsight
38.Nicholson  K, Stewart  M, Thind  A.  Examining the symptom of fatigue in primary care: a comparative study using electronic medical records.   J Innov Health Inform. 2015;22(1):235-243. doi:10.14236/jhi.v22i1.91PubMedGoogle ScholarCrossref
39.Morgan  S, Henderson  KM, Tapley  A,  et al.  Investigation of fatigue by Australian general practice registrars: a cross-sectional study.   J Prim Health Care. 2015;7(2):109-116. doi:10.1071/HC15109PubMedGoogle ScholarCrossref
40.Delpech  R, Bloy  G, Panjo  H, Falcoff  H, Ringa  V, Rigal  L.  Physicians’ preventive practices: more frequently performed for male patients and by female physicians.   BMC Health Serv Res. 2020;20(1):331. doi:10.1186/s12913-020-05136-2PubMedGoogle ScholarCrossref
41.Krähenmann-Müller  S, Virgini  VS, Blum  MR,  et al.  Patient and physician gender concordance in preventive care in university primary care settings.   Prev Med. 2014;67:242-247. doi:10.1016/j.ypmed.2014.08.004PubMedGoogle ScholarCrossref
42.Le  JV, Pedersen  LB, Riisgaard  H,  et al.  Variation in general practitioners’ information-seeking behaviour—a cross-sectional study on the influence of gender, age and practice form.   Scand J Prim Health Care. 2016;34(4):327-335. doi:10.1080/02813432.2016.1249057PubMedGoogle ScholarCrossref
43.Salinas  M, López-Garrigós  M, Díaz  J,  et al.  Regional variations in test requiring patterns of general practitioners in Spain.   Ups J Med Sci. 2011;116(4):247-251. doi:10.3109/03009734.2011.606927PubMedGoogle ScholarCrossref
44.Read  AJ, Waljee  AK, Sussman  JB,  et al.  Testing practices, interpretation, and diagnostic evaluation of iron deficiency anemia by US primary care physicians.   JAMA Netw Open. 2021;4(10):e2127827. doi:10.1001/jamanetworkopen.2021.27827
ArticlePubMedGoogle ScholarCrossref
45.Bayen  S, Le Grand  C, Bayen  M, Richard  F, Messaadi  N.  Anemia management in non-menopausal women in a primary care setting: a prospective evaluation of clinical practice.   BMC Fam Pract. 2020;21(1):13. doi:10.1186/s12875-020-1086-5